pylimma.plot_sa

pylimma.plot_sa(fit, xlab='Average log-expression', ylab='sqrt(sigma)', zero_weights=False, pch='o', cex=0.3, col=('black', 'red'), ax=None, **kwargs)

Sigma vs Amean plot. Port of R limma’s plotSA.

The y-axis is sqrt(sigma). When s2_prior is an ndarray, the variance-trend overlay is sqrt(sqrt(s2_prior[order(x)])) in sorted x order. When s2_prior is scalar, a single horizontal line is drawn at sqrt(sqrt(s2_prior)).

Under robust eBayes (per-gene df_prior), outlier genes are identified via p_adjust(2 * pmin(pdn, pup), method='BH') <= 0.5 and plotted in col[1].

Parameters:

• xlab (str)

• ylab (str)

• zero_weights (bool)

• cex (float)

• col (tuple[str, str])